SauerkrautLM-Gemma-7B: benchmark results
VAGOsolutions and Hyperspace.ai's SauerkrautLM-Gemma-7b (alpha), an early-stage Gemma 7B fine-tune using SFT, DPO and a novel laser-QLoRA partial-freezing technique from LaserRMT research. Provider: VAGOsolutions. Released 2024-02-22. Access: Open.
Aggregate Rating 1385 ± 23, rank #1249 of 1635 rated models, from 29 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Medical LLM - PubMedQA | 75.8 | Accuracy (%) | 79.2 |
| Open LLM Leaderboard v1 - GSM8K | 63.68 | Accuracy (%) (5-shot) | 77.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 61 | MC2 (%) (0-shot) | 76.9 |
| Open Portuguese LLM - ENEM | 68.58 | Accuracy (%) | 70.5 |
| Open Portuguese LLM - BLUEX | 56.33 | Accuracy (%) | 65.9 |
| Open LLM Leaderboard v1 - MMLU | 63.76 | Accuracy (%) (5-shot) | 63.1 |
| Open Portuguese LLM - OAB Exams | 45.74 | Accuracy (%) | 62.2 |
| Open Portuguese LLM - FaQuAD NLI | 65.7 | Macro F1 (%) | 51 |
| Open LLM Leaderboard v1 - WinoGrande | 76.64 | Accuracy (%) (5-shot) | 49.6 |
| Open LLM Leaderboard v1 - HellaSwag | 81.91 | Normalized accuracy (%) (10-shot) | 47.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 59.98 | Normalized accuracy (%) (25-shot) | 45.8 |
| Open Medical LLM - MMLU College Biology | 72.92 | Accuracy (%) | 44.3 |
Interactive version: theaggregate.ai/model?slug=sauerkrautlm-gemma-7b · How It Works · Data refreshed daily, snapshot 2026-11-05.